Spatial Distribution and Implementation of the K-Means Clustering Method at Hotspots in North Sumatra

Abstract

Hotspots are indicators of forest and land fires. Hotspot monitoring can be carried out with the help of remote sensing tools and geographic information systems. Hotspot data is obtained from the MODIS sensors from the TERRA and AQUA satellites which contain information on latitude and longitude coordinates and the level of confidence divided by three levels, namely low, medium and high confidence levels. Based on the spatial results, the number of hotspots in North Sumatra Regency is in February, March, June, July, and August. Districts that are dominant with hotspots are Karo Regency, Labuhan Batu Regency, Mandailing Natal Regency, Padang Lawas Regency and South Tapanuli Regency. Based on the results, the process of applying the k-means clustering method to the weka application, the data obtained is in the form of a clustered group and the results can be made into indicators in determining hotspots in districts in North Sumatra province per month.